The stem cell model for Acute Myelogenous Leukemia proposes that a leukemia stem cell population, LSC, is responsible for maintaining the disease. Properties that distinguish LSCs from their normal counterparts have been proposed and may represent targets for novel therapies for this deadly disease. There is little evidence that the model is clinically relevant apart from studies that demonstrated that the risk of relapse is related to the frequency of leukemic CD45+CD34+CD38low cells in a patient remission marrow. The goal of our research is to unambiguously demonstrate the clinical relevance of the leukemic stem cell model. As a step towards achieving this goal, we propose to 1.) isolate and functionally validate the LSC populations from patients undergoing treatment for AML at time of diagnosis and relapse 2.) to demonstrate the stability of the LSC phenotype by directly comparing the surface antigen and molecular signaling profiles of pre-therapy and relapse LSCs and, 3.) to determine if previously reported LSC specific properties are maintained in the remission state. We are accruing patients to a multi-center effort to accomplish this goal. We have characterized the surface antigen expression of normal and leukemic CD45+CD34+CD38low cells and identified differentially expressed surface markers that allow the isolation of leukemic population during the remission state. We will analyze candidate populations from pre-therapy and post-relapse samples and perform NOD/SCID engraftment assays to identify patient specific LSCs. We will then isolate and characterize the LSC populations with regard to dysregulation of signaling and activation of NFKB. We will then utilize the patient specific LSC surface antigen profile to isolate LSCs from the patients'remission samples and confirm or refute our hypothesis that LSC specific properties are stable throughout the clinical course. Data obtained from this study will provide direct evidence of the clinical relevance of LSCs and the potential of targeting them. PUBLIC HEALTH RELEVANCE: The cancer stem cell model for malignancy proposes that a small population of cancer cells initiate and maintain tumors and represent a therapy refractory resevoir for relapse. This multi-center effort represents the first prospective effort to assess the degree of inter-patient variability and phenotypic stability of an individual patient's cancer stem cells during therapy. The results from the proposed studies will represent a strong first step in verifying the importance/relevance of targeting cancer stem cells in improving outcome for patients with cancer.